Training complete

README.md

@@ -19,14 +19,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Salesforce/codet5-base](https://huggingface.co/Salesforce/codet5-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.7440
-- Rouge1: 0.0524
-- Rouge2: 0.0010
-- Rougel: 0.0536
-- Bleu: 0.1380
-- Meteor: 0.0295
-- Codebleu: 0.0247
-- Em: 0.0
+- Loss: 0.3737
+- Rouge1: 0.4604
+- Rouge2: 0.2526
+- Rougel: 0.4357
+- Bleu: 19.1354
+- Meteor: 0.5043
+- Codebleu: 0.2425
+- Em: 0.07
 
 ## Model description
 
@@ -49,8 +49,8 @@ The following hyperparameters were used during training:
 - train_batch_size: 4
 - eval_batch_size: 4
 - seed: 42
-- gradient_accumulation_steps: 16
-- total_train_batch_size: 64
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 3

runs/Aug05_11-13-14_cb69df8c9ea7/events.out.tfevents.1722856395.cb69df8c9ea7.4878.13

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8a0a623ce66a4ee6b784ad7edd378530f8ab241ce082626fb313bcbc69e471a
-size 6034
+oid sha256:51bab6a96e1c275be9637506d1a54c28cd62065efec92934ab36f8f805a1dc81
+size 6382

runs/Aug05_11-13-14_cb69df8c9ea7/events.out.tfevents.1722857220.cb69df8c9ea7.4878.14

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a44ab9198c21d8275fc3229d8f0f8bcbe0ccfef325994d77ad9831e64fc2e2f4
+size 693